ABSTRACT
Objective To compare the effect of ARIMA models with and without covariates in predicting the number of HIV infections among young students in Dalian. Methods First, univariate correlation analysis was performed on the network, STD sequence and HIV sequence to understand whether there was a correlation and lag relationship between them. Secondly, variables with the strongest correlation and predictive value and HIV infection numbers were used as the baseline data to establish an ARIMA model with covariates and a general ARIMA model without covariates, and to predict the HIV number from 2019 to 2021. The average absolute errors were used as evaluation indexes to compare the prediction effects of the two models. Results A total of 841 cases of HIV infection among young students were reported in Dalian from 2013 to 2021. The results of univariate correlation analysis showed that the search index of the keyword AIDS in the Baidu Index in a given month from 2013 to 2019 was significantly positively correlated with the number of HIV infections in that month (r=0.302, P=0.006), and gonorrhea was negatively correlated with the number of HIV infections with a lag of 2 months (r=-0.250, P =0.024). Using gonorrhea incidence number and HIV infection number as the basic data, an ARIMA model with covariates and a general ARIMA model without covariates were established to predict the number of HIV infection among young students from 2019 to 2021, and the average absolute errors were 17.621% and 66.17%, respectively. Conclusion Compared with the general ARIMA model without covariates, the ARIMA model based on the combined use of STD incidence and HIV infection is more suitable for predicting the number of HIV infections among young students in Dalian, but the average absolute error of the model is still large, which needs further improvement in the future research.
ABSTRACT
Objective:To analyze the epidemiological trends and population characteristics of inflammatory bowel disease(IBD)in China by searching keywords related to IBD through Google Trends, Baidu index, and WeChat index, so as to provide a reference for national epidemiological studies on IBD.Methods:IBD-related hot words such as "inflammatory bowel disease" , "Crohn′s disease" , "ulcerative colitis" , "gastroesophageal reflux disease (GERD)" , "irritable bowel syndrome (IBS)" and " Helicobacter pylori ( H. pylori)" were selected. The search volume and trends of the above keywords in the world and China were analyzed through Google Trends, Baidu index and WeChat index. The epidemiological characteristics of IBD in China were summarized. Descriptive methods were used for statistical analysis. Results:The results of Google Trends analysis showed that among 5 common digestive diseases (GERD, IBS, H. pylori infection, IBD and peptic ulcer), GERD was the most concerned disease, while IBD was not the focus among the common digestive diseases. When the global searching scope was limited to IBD related hot words, Crohn′s disease was the disease of primary concern among IBD-related diseases. In South America, South Asia, and parts of Africa, ulcerative colitis was mainly concerned, and in China and countries of Southeast Asia, IBD was more concerned. The searching results of Baidu index indicated that among the national search for IBD-related hot words, the 3 keywords of "inflammatory bowel disease" , "ulcerative colitis" and "Crohn′s disease" were all highly searched, the overall daily average of the search indexes of the 3 keywords were 325, 1 320 and 2 559, respectively, and the searching volume of "Crohn′s disease" was higher than "inflammatory bowel disease" and "ulcerative colitis" . The national wide trends of search volume for "inflammatory bowel disease" , "ulcerative colitis" and "Crohn′s disease" were similar, the search heat gradually decreased from the east coast to the northwest of China, which basically coincided with the three-level ladder trend of China′s economic development, suggesting that the level of economic development was related to the incidence of IBD. The results of Baidu index analysis showed that the main populations who searched IBD-related keywords were young adults and women aged from 20 to 39 years old. The results of WeChat index analysis revealed that the searching volume of "inflammatory bowel disease" , "ulcerative colitis" and "Crohn′s disease" were 205 000, 195 000 and 120 000, respectively, and the search volume for " inflammatory bowel disease" was the highest. The official account (90.27%) and the video account (7.43%) occupied the main sources of IBD-related information on mobile terminals. Conclusions:The IBD-associated internet activities reveal a global lack of public awareness of IBD. China also faces the same problem. The search trend is consistent with the epidemiology of IBD, which may be helpful for future epidemiological research of IBD in China. Mobile media will be a potential force in promoting the patient education and disease management of IBD in China.
ABSTRACT
Objective:To explore the hotspots of the public concerns about quitting smoking before and during the coronavirus disease 2019 (COVID-19) pandemic.Methods:The monthly search index data of four key words, “the harm of smoking”，“quitting smoking”，“passive smoking” and “electronic cigarette” before the COVID-19 (from January, 2016 to December, 2019) and during the COVID-19 (from January, 2020 to December, 2020) pandemic were manually collected from the Baidu Index platform. The one-way ANOVA was used to compare the differences among annual search indexes of four keywords. The linear regression model was used to test the annual linear trend of search index of each keyword before the COVID-19, and the one-sample t-test was used to analyze the differences of the search indexes during the COVID-19 and the means before the COVID-19. Results:During 2016—2020, the annual average search index of “electronic cigarette” was the highest (1.465 million), followed by “the harm of smoking” (0.884 million) and “quitting smoking” (0.780 million), while “passive smoking” was the lowest (0.171 million). Before the COVID-19 (2016—2019), the search index of each keyword had positive cumulative growth, but there was no significant linear trend of annual change ( P>0.05). During the COVID-19 period (2020), the search index of “the harm of smoking” decreased significantly (0.794 million vs 0.907 million, P=0.011), and “quitting smoking” (0.591 million vs 0.827 million, P=0.172) and “passive smoking” (0.164 million vs 0.172 million, P=0.257) showed a downward trend, while the search index of “electronic cigarettes” bucked the trend and increased significantly (1.825 million vs 1.375 million, P=0.010). Conclusions:The public′s online attention to “electronic cigarette” increases dramatically during COVID-19 along with other information about quitting smoking decreases. There are urgent needs for strengthening regulation and monitoring of electronic cigarettes and carrying out more effective publicity of scientific methods to help quit smoking.
ABSTRACT
Objective@#To establish a prediction model of herpangina epidemic trend based on Baidu index and hand, foot and mouth disease, so as to provide insights into analyses of communicable disease epidemics with limited or missing surveillance data.@*Methods@#The incidence of hand, foot and mouth disease in Zhejiang Province during the period from the first week of 2015 through the 39th week of 2021 was retrieved from the China Information System for Disease Control and Prevention, and the Baidu index of hand, foot and mouth disease and herpangina was collected via the Baidu search engine during the same period. The correlation between the Baidu index and time series of hand, foot and mouth disease was examined using wavelet analysis. In addition, a random forest training model was created based on the Baidu index and incidence of hand, foot and mouth disease, and the fitting effectiveness was evaluated using the mean percentage error, while the Baidu index of herpangina was included in the model to predict the epidemic trend of herpangina during the study period.@*Results@#The Baidu index of herpangina and hand, foot and mouth disease, and the Baidu index and incidence of hand, foot and mouth disease all appeared two peaks at the 26th and 52th week. The phase difference was less than 0.1 week between the Baidu index and time series of hand, foot and mouth disease, and the mean percentage error of the training model was 13.07%, with high concordance between the predicted number and actual report number of cases with hand, foot and mouth disease. The numbers of herpangina cases were predicted to be 28 822, 27 341, 28 422, 51 782, 52 457 and 5 691 from 2015 to 2020, and there were totally 48 702 herpangina cases reported until the 39th week of 2021. Like hand, foot and mouth disease, the incidence of herpangina peaked between May and July.@*Conclusion@#The random forest training model based on the Baidu index and incidence of hand, foot and mouth disease is feasible to predict the epidemic trend of herpangina.
ABSTRACT
Objective@#To explore the temporal and spatial characteristics of search volume (SV) of allergic rhinitis (AR) and related search terms on Baidu Index, as well as the correlation between SV from Baidu Index and epidemiology of AR.@*Methods@#SV of AR and related search terms in mainland area of China from January 2012 to December 2016 were extracted from Baidu Index. Firstly, we investigated the temporal and spatial characteristics of SV of AR and related search terms ("pollen allergy" and" dust mite allergy", etc.), and explored the correlation between SV of AR and SV of related search terms. Secondly, the SV of specific search terms in Beijing and Guangzhou were compared with the actual pollen count in Beijing from April to September in 2016 and the monthly outpatient volume of AR patients in the Otorhinolaryngology Hospital, The First Affiliated Hospital of Sun Yat-sen University from 2014 to 2016 respectively to analyze the consistency between SV from Baidu Index and the real-world data. Spearman correlation analysis was used to explore the correlation between different search terms according to the normality and homogeneity of variance of the data. P<0.05 was considered statistically significant.@*Results@#SV of AR showed repetitively seasonal pattern, with the first peak in May and the second peak from August to September. SV of AR in northeast China, north China and northwest China were correlated with both SV of "pollen allergy" and "dust mite allergy" (r value was 0.533, 0.638, 0.609, 0.791, 0.517, 0.682, respectively, all P<0.05). In east China, central China, south China and southwest China, SV of AR showed significant correlation only with SV of "dust mite allergy" (r value was 0.897, 0.884, 0.799, 0.937, respectively, all P<0.05), but not with SV of "pollen allergy" (r value was 0.110, 0.171, 0.020, 0.265, respectively, all P>0.05). The correlation between pollen count and the SV of AR on the day and SV of "pollen allergy" on the next day was the highest (r value was 0.692 and 0.713, respectively, all P<0.05). The SV of AR was correlated with the monthly outpatient volume of AR patients one month and two months later (r value was 0.523 and 0.503, respectively, all P<0.05).@*Conclusion@#The SV of AR and related search terms show different trends in different times and regions which are consistent with the AR epidemiological survey, and significantly correlate with pollen count and AR outpatient volume.
ABSTRACT
Objective To investigate network user's attention rate of the keyword“cardiovascular and cerebrovascular diseases”(CCD). Methods The keyword was analyzed by using Baidu index. Results The overall trend of keyword search rose slightly. The mobile search grew significantly. During the Spring Festival overall search decreased significantly. According to clustering analysis the word“symptom”was most often used with“CCD”;30 to 39 years old age group accounted for about 50%of total number of online search. Men accounted for 70% of all keyword search, women accounted for 30%. The rankings of CCD search number were more on top in the economically developed area. Conclusion It is necessary to strengthen publicity and education to prevent the happening of CCD during Spring Festival. Men need to raise awareness about CCD. The construction of rural informatization should be speeded up. Network education could be used as a part of method for the primary and secondary prevention of CCD. Keyword analysis method could effectively analyze the network user search behavior and find information they need.